The PR author made changes! So at least it's not a "slop and forget" change.
However, all of their responses have that horrific obsequious tone of LLMs.
And one of my comments was "this call to openFile() seems to render the file, which would be inefficient for multiple files - can you just add it to the database instead?" and their response was "Good point, I've fixed it to be less complicated and more efficient"…
…and then they simplified the OTHER function (which did have more complex async for unexplained reasons)… and it STILL uses the function that I said was inefficient 🙄
Meanwhile, the comments on "this change seems unrelated" got "you're right, that was included by accident, I've reverted it".
I'm calling slop, and I hate the state of the world.
I've just read a PR on a FOSS project and… I think it's AI slop 😐
There's nothing much in the code that's blatant slop. It looks reasonable, give or take a weird choice on one function to call. But it's also a function that a human might naively decide to call.
But then there's the PR text. It has footnote-style links. And it says generic and unhelpful things like "swapped X for Y and updated Z"… Yes… I can see that from the diff. But WHY? 🙄
IBBoard